Invasiveness in tumors is an initial and essential step in the development of malignancy and the principal expression of malignancy in primary intracranial tumors, but little is known about the cellular and molecular factors that lead to its development. This project is a study of intracellular and cell surface factors which effect the development of invasiveness in tumors. We have developed an experimental model for the production of primary intracranial tumors that vary predictably in their invasiveness for brain. The most invasive tumors morphologically exhibit substrate dependent inhibition of locomotion and have viral DNA integration patterns indicative of genomic selection and rearrangement with passage. When passed and cloned for anchorage dependence, they remain invasive for brain and histologically resemble human malignant astrocytic series tumors. Using this model, we propose to: 1. define more precisely, using cloned and subcloned tumor cells, the role of anchorage dependence in the development of invasiveness. Specifically, we shall determine a) the efficiency of plating and growth in soft agar using standard and reduced concentrations of cells; b) the effects of growth in medium with reduced serum concentration; and c) the stability of the newly acquired invasive phenotype with further passage and/or cloning. 2. determine whether the patterns of expression of cell surface glycoproteins and SV40 viral-encoded or induced intracellular or cell surface antigens specifically alter cell-substrate attachment and thereby directly, and perhaps causally, affect the development of tumor invasiveness. Specifically, we propose to a) determine, using double staining indirect immunofluorescence techniques, whether localization of fibronectin or laminin to the attachment sites of our invasive glial fibrillary acidic protein-containing cells is altered or enhanced and whether such cells exhibit increased adherence to exogenous glycoprotein; b) establish, using immunoprecipitation and cell surface labeling techniques and antisera specifically prepared against invasive and noninvasive Cx tumor cell antigens, whether antigens unique to, or greatly amplified in, the invasive cell are characteristically and reproducibly present; and c) prepare monoclonal antisera against such cellular and viral-encoded or induced antigens. These studies should have important therapeutic implications.